Detection of eye contact with deep neural networks is as accurate as human experts.

Eye contact is among the most primary means of social communication used by humans. Quantification of eye contact is valuable as a part of the analysis of social roles and communication skills, and for clinical screening. Estimating a subject's looking direction is a challenging task, but eye contact can be effectively captured by a wearable point-of-view camera which provides a unique viewpoint. While moments of eye contact from this viewpoint can be hand-coded, such a process tends to be laborious and subjective. In this work, we develop a deep neural network model to automatically detect eye contact in egocentric video. It is the first to achieve accuracy equivalent to that of human experts. We train a deep convolutional network using a dataset of 4,339,879 annotated images, consisting of 103 subjects with diverse demographic backgrounds. 57 subjects have a diagnosis of Autism Spectrum Disorder. The network achieves overall precision of 0.936 and recall of 0.943 on 18 validation subjects, and its performance is on par with 10 trained human coders with a mean precision 0.918 and recall 0.946. Our method will be instrumental in gaze behavior analysis by serving as a scalable, objective, and accessible tool for clinicians and researchers.

Increased Eye Contact During Conversation Compared to Play in Children With Autism.

Children with autism have atypical gaze behavior but it is unknown whether gaze differs during distinct types of reciprocal interactions. Typically developing children (N = 20) and children with autism (N = 20) (4-13 years) made similar amounts of eye contact with an examiner during a conversation. Surprisingly, there was minimal eye contact during interactive play in both groups. Gaze behavior was stable across 8 weeks in children with autism (N = 15). Lastly, gaze behavior during conversation but not play was associated with autism social affect severity scores (ADOS CSS SA) and the Social Responsiveness Scale (SRS-2). Together findings suggests that eye contact in typical and atypical development is influenced by subtle changes in context, which has implications for optimizing assessments of social communication skills.

Swab transport in Amies gel followed by frozen storage in skim milk tryptone glucose glycerol broth (STGGB) for studies of respiratory bacterial pathogens.

Nasopharyngeal carriage studies are needed to monitor changes in important bacterial pathogens in response to vaccination and antibiotics. Commercial swab transport followed by transfer to skim milk tryptone glucose glycerol broth for frozen storage is an option for studies of Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis.

Topical ciprofloxin versus topical framycetin-gramicidin-dexamethasone in Australian aboriginal children with recently treated chronic suppurative otitis media: a randomized controlled trial.

BACKGROUND: Chronic suppurative otitis media (CSOM) affects many children in disadvantaged populations. The most appropriate topical antibiotic treatment in children with persistent disease is unclear. METHODS: Children with CSOM despite standard topical treatment were randomized to 6-8 weeks of topical ciprofloxacin (CIP) versus topical framycetin-gramicidin-dexamethasone (FGD). Otoscopic, audiologic, and microbiologic outcomes were measured using standardized assessments and blinding. RESULTS: Ninety-seven children were randomized. Ear discharge failed to resolve at the end of therapy in 70% children regardless of allocation [risk difference = -2%; (95% CI: -20 to 16)]. Healing of the tympanic membrane occurred in one of 50 children in the CIP group and none of 47 children in the FGD group. Severity of discharge failed to improve in more than 50% children in each group, and mean hearing threshold (38 dB and 35 dB) and proportion of children with greater than 25 dB hearing loss (98% and 88%) were not significantly different between the CIP and FGD groups. Side effects were rare. CONCLUSIONS: This study showed a similarly low rate of improvement or cure in children with persistent CSOM for both CIP and FGD topical therapies. Complications and side effects were insufficient to cease therapy or inform prescribing of either therapy.

Comparison of nasal swabs with nose blowing for community-based pneumococcal surveillance of healthy children.

The nasopharynx (NP) is the preferred site for detection of Streptococcus pneumoniae in young children, but NP sampling is not well tolerated. We compared nose blowing with paired nasal swabs. The sensitivity of nose blowing was 46% (95% confidence interval [CI] 38 to 56%), which increased to 94% (95% CI, 85 to 98%) for children with visible secretions.

The clinical course of acute otitis media in high-risk Australian Aboriginal children: a longitudinal study.

BACKGROUND: It is unclear why some children with acute otitis media (AOM) have poor outcomes. Our aim was to describe the clinical course of AOM and the associated bacterial nasopharyngeal colonisation in a high-risk population of Australian Aboriginal children. METHODS: We examined Aboriginal children younger than eight years who had a clinical diagnosis of AOM. Pneumatic otoscopy and video-otoscopy of the tympanic membrane (TM) and tympanometry was done every weekday if possible. We followed children for either two weeks (AOM without perforation), or three weeks (AOM with perforation), or for longer periods if the infection persisted. Nasopharyngeal swabs were taken at study entry and then weekly. RESULTS: We enrolled 31 children and conducted a total of 219 assessments. Most children had bulging of the TM or recent middle ear discharge at diagnosis. Persistent signs of suppurative OM (without ear pain) were present in most children 7 days (23/30, 77%), and 14 days (20/26, 77%) later. Episodes of AOM did not usually have a sudden onset or short duration. Six of the 14 children with fresh discharge in their ear canal had an intact or functionally intact TM. Perforation size generally remained very small (<2% of the TM). Healing followed by re-perforation was common. Ninety-three nasophyngeal swabs were taken. Most swabs cultured Streptococcus pneumoniae (82%), Haemophilus influenzae (71%), and Moraxella catarrhalis (95%); 63% of swabs cultured all three pathogens. CONCLUSION: In this high-risk population, AOM was generally painless and persistent. These infections were associated with persistent bacterial colonisation of the nasopharynx and any benefits of antibiotics were modest at best. Systematic follow up with careful examination and review of treatment are required and clinical resolution cannot be assumed.

Streptococcus pneumoniae and noncapsular Haemophilus influenzae nasal carriage and hand contamination in children: a comparison of two populations at risk of otitis media.

BACKGROUND: Australian Indigenous children living in remote areas have rates of tympanic membrane perforation as high as 60%, almost 100 times the prevalence in urban child care settings (<1%). Relative rates of pneumococcal nasal carriage do not reflect this difference in disease risk. METHODS: Cross-sectional comparison of nasal carriage and hand contamination in children younger than 4 years of age from urban child-care centers and Indigenous children 3-7 years of age from a remote community. Almost identical methods of nasal swab collection, transport and culture were used. Data on pneumococcal antimicrobial susceptibility patterns and serotypes are also reported. RESULTS: For Indigenous children compared with children in child care, the relative risk of nasal carriage of either pneumococcus or noncapsular Haemophilus influenzae was <2-fold [relative risk, 1.7; 95% confidence interval (CI), 1.5, 1.9], the risk of simultaneous nasal carriage was almost 3-fold (78% versus 28%; relative risk, 2.9; 95% CI 2.3, 3.5), and the risk of pneumococcal hand contamination was 8-fold higher (37% versus 4%; relative risk, 8.4; 95% CI 4.6, 15.2). For simultaneous hand contamination, the risk was 23-fold (8% versus 0.3%; relative risk, 23.1; 95% CI 2.9, 185.4). Remote Indigenous children also had a more diverse serotype distribution (25 versus 14 serotypes identified). CONCLUSIONS: Simultaneous nasal carriage of Streptococcus pneumoniae and H. influenzae and hand contamination are simple indicators of risk for use in studies of otitis media in populations at risk for tympanic membrane perforation.